skip to Main Content

An excellent article in the Journal of the American Geriatrics Society provides insights into primary care physicians views of dementia care. Since the vast majority of patients with dementia are cared for by primary care providers, improving care makes it very important to understand the perspective of their providers. The paper is authored by Dr. Dorothy Harris of UCLA.

The authors surveyed 164 primary care providers (mostly family physicians and internists). The survey primarily compared the providers’ views of caring for dementia, with that of caring for diabetes and heart disease. The key findings:

  • Providers were much more likely to somewhat or strongly agree that dementia is difficult to manage (56%) than heart disease (22%) or diabetes (22%)
  • Providers were much less likely to stronlgy agree that they could improve the quality of life for patients with dementia (31%) compared to heart disease (59%) or diabetes (62%)
  • Providers were much less likely to strongly agree that their health care organization had the resources to manage dementia (21%) than for heart disease (52%) or diabetes (49%).

On the one hand, it is not surprising that clinicians lacked confidence that they could improve the quality of life for patients with dementia. For heart disease and diabetes, there are effective pharmacologic therapies that significantly alter the natural history of the disease and clearly improve outcomes and quality of life. For dementia, the ability to change the natural history of the disease is very limited, and existing pharmacologic treatments are of marginal benefit.

On the other hand, behavioral and psychosocial interventions are very effective at improving patient and caregiver quality of life and outcomes in dementia. (see Vickrey, Belle, and Callahan studies). However, these interventions take training to administer, and require team based approaches. They are not purely clinic based, but require collaboration with community-based service providers. Further, Medicare and other insurers do not pay for these approaches and support for services aimed at helping family caregivers is woefully inadequate. Bree Johnston and I previously notedthat if these behavioral interventions were drugs, they would be on the fast track to approval.

The difficulty reported by clinicians likely reflects the reality that few health systems have the necessary systems in place to optimally care for patients with dementia. Improving care will require changes at the system level that enable primary care providers to coordinate optimal care. And as noted by Harris:

“The challange lies in instituting cultural changes in primary care practice and training—such as making the caregiver a focal point of care interventions–to elevate this type of care to the same status as practices for treating heart disease, diabetes mellitus, and other common conditions that rely more centrally on prescribing medications.”

This Post Has 5 Comments

  1. As usual, very thoughtful, Ken.

    The observation that a health service, such as GRACE or the IMPACT primary care collaborative depression treatment model would have become a covered benefit with a payment attached, if they were a pill, device, or conventional procedure (i.e., surgical) is one we have toyed with for years.

    Other parties have gone as far as discussing with CMS what becoming a covered benefit would take. In public they say and, indeed, the regs suggest that this is possible. In private, CMS staff have said not even to try. It would just be perceived as irritating and lead to no good outcome.

    One problem is that parts of the services ARE already paid for under E&M billing. There is some contrary thing in me, that makes me point out whenever someone writes "Medicare and other insurers do not pay for these approaches" that CMS payments already assume a fair amount of patient education, follow-up phone calls, and so on, which is built into the rates for the E&M payments.

    What we really mean when we say "they won't pay" is that there isn't a piece work, more-for-more, fee-for-service payment attached. And probably, all things considered, there shouldn't be. Delivering an intervention like GRACE or IMPACT in an effective way is much harder than choosing the right pill – not that perscribing always goes so well either.

    Really some kind of pay-for-performance, at-risk payments, or other more global payment approach is needed. But those kinds of changes are scary and have all kinds of possible unintended side effects, as GeriPal has been discussing for the last few weeks.

  2. Dementia is interesting. While I was working at a hospice in town, we often struggled with dementia patients getting admitted "too early." Many of these patients and their families had goals that were consistent with hospice but their bodies did not yet meet the criteria for hospice. We had to discharge these patients given how strict the fiscal intermediary had become regarding these patients. It always felt terrible because we knew these patients were heading back into the dangerous, acute care oriented, fee-for-service system that wasn't designed for them.

    There are very few conditions like dementia where the goals align with hospice before the body does. Indeed, in cancer, the patient's body has often be aligned with hospice for 5.75 months before their goals align with hospice.

    What is my point? I agree with Chris Langston that another "more for more" FFS proposal is likely not appropriate. However, until we have some sort of payment reform, what if there was a per diem payment for care coordination and support for dementia patients and their families. It would require some research but I imagine the costs of such a program would be offset by delays in institutionalization and decreases in hospitalizations.

  3. Here's another article on palliative care and dementia from The Lancet Neurology.
    "It is, perhaps, surprising that a paper published in 2009 on the symptoms seen in advanced dementia should warrant much comment. After all, dementia has been around for a very long time. And, yet, a recent paper by Mitchell and colleagues is timely and important. Why?
    This study is a timely step in terms of pushing forward the important role of palliative care in dementia. In 323 residents with dementia in nursing homes in Boston, USA, 177 (54·8%) died over the course of 18 months."

  4. Thanks to all for their comments. A few additional thoughts:

    Chris–thank you for sharing your perspective on CMS and payment for these integrated service interventions. It does make me realize that the statements as to what are and are not paid for are more complex than I am making them out to be.

    I can understand in a very superficial sense how CMS can argue that some of these services are paid for because they are already bundled into the E/M payments. But, primary care is not well reimbursed as it is. Asking primary care providers to do even more uncompensated service integration work would further exacerabate this problem.

    And on a deeper level, I think the CMS argument is actually not accurate. My understanding is that the current reimbursement system is based on studies of the time and intensity required to deliver different medical services. Many feel these studies were heavily biased against cognitive services to begin with. Certainly, these studies did not consider the time requirements and care intensity required to manage an intervention such as IMPACT or the dementia interventions I cited. In fact, it is hard to imagine how a practice that spends much effort in these areas can be financially viable.

    Your point that the answer probably lies in more major structural reform, where there are $ incentives for good outcomes seems right on. Still, it seems it would be good to do something to support these services while we are waiting for this to happen.

    You and some of other readers know much more about payment structures than I do, and I would welcome further education and insights.

    Dan—your comments about dementia are very interesting. It makes me think about something I have heard Sean Morrison emphasize a number of times—That the needs for palliative care services often are not dictated by prognosis—and that eligibility for these services should be driven by clinical need rather than prognosis. It sure seems dementia is a classic case where palliative needs and prognosis are often poorly correlated.

    Jan–thank you for noting Dr. Corn's excellent essay. It makes a lot of great points in a short amount of space. Perhaps we can get one of our bloggers to do a full post on it.

Leave a Reply

Your email address will not be published. Required fields are marked *

Back To Top